Cardiovascular disease (CVD) risk estimation has focused on outpatient data from whites according to age, sex, LDL-cholesterol (LDL-C), HDL-cholesterol (HDL-C), diabetes, smoking, and blood pressure (BP) information. Risk factors (RFs) and overall CVD risk are associated with genetic variations in genome-wide association studies (GWAS), and phenotypes have largely been based on single RF measurements using a Framingham approach. Previous research has focused on European Americans (EA), and generally has not included Veterans. There is little information on CVD risk factor genes in African Americans (AA) or Hispanic Americans (HA), two population groups that are extremely important in the VA. With the availability of the Million Veteran Program (MVP) cohort, we have a unique opportunity to study genes and CVD risk among these subgroups. Up to now methods have not been developed to link MVP questionnaire and genetic data to the Veteran's electronic healthcare information. We propose to address scientific gaps by focusing on multiple ethnicities, rare variants, and antecedent risk factor levels using the MVP cohort. In the proposed study, we first create a virtual baseline exam by developing methods to link data from MVP baseline examination to the VA electronic health record using a systematic approach. Results for Veterans are influenced by geographic region in the U.S., race/ethnicity, frequency of outpatient clinic visits, and other variables. Based on our systematic approach of handling MVP questionnaire data and the electronic healthcare data, we assess the genome-wide associations of both common and rare alleles with CVD RFs in causal pathways, with consideration of the environment (diet quality, pharmacological treatment), and assessment of current and antecedent RF levels. Methods include single VA outpatient measurements of quantitative CVD RFs (LDL-C, non- HDL-C, triglycerides, body mass index) with and without diet quality adjustment using the Willett Food Frequency Questionnaire performed at the MVP baseline visit. Other methods include pharmacologic treatment of CVD RFs to derive imputed untreated RF levels, and antecedent quantitative CVD RFs measured at VA outpatient visits 5 and 10 years before the MVP baseline visit when such data are available. We will perform common variant association studies (CVAS) and rare variant association studies (RVAS) testing the association of genetic variants to quantitative CVD risk for a) prevalent coronary heart disease (CHD) [myocardial infarction, history of coronary bypass grafting (CABG), history of percutaneous coronary intervention (PCI)] and b) prevalent atherothrombotic stroke, with comparison of effects by race and ethnicity, and c) examine the multigenic association of CHD and stroke using the genetic risk score (GRS) of validated CVD-associated SNPs within and across ethnicity. This project will provide a platform for CVD incidence analyses for MVP participants across the VA in the future. Furthermore, the proposed study findings will allow for the comparison of the impact of genetic variants on cardio metabolic RFs risk factors and atherosclerotic disease prevalence across AA, HA, and EA Veterans. CVD accounts for an extremely large health care burden in veterans. This project will investigate the role of genetic and environmental determinants of CVD risk factors in MVP participants across different ethnic and racial groups.